of the clock, so that it was thought advisable to discard the first pair of observations, 

 especially as they agreed in no way with the later ones. 



The observations for clock rate of " 991 " were made by Mr. Merfield with the 

 8-inch meridian circle. 



The collimation was determined by two collimators of 6-inch aperture and 78-inch 

 focal length, placed on opposite sides of the rotation axis of the telescope. The level 

 correction was found by coincidence of the middle wire with its reflection in a dish 

 of mercury below the floor. Bessel's method was used by Mr. Merfield for reduction of 

 the observations. 



The errors of the instruments were determined by Mr. Merfield for each set of stars. 

 The programme included at least six clock stars and four circumpolar stars. The 

 positions of the clock stars are taken from the " Nautical Almanac." 



In Table LIV are found the Instrumental Constants and rate of No. 991. 



TABLE LIV. 



Errors and Rates of Transit Clock (Frodsham No. 991) determined from observations 

 made with the 8-inch meridian circle instrument by C. ,1. Merfield. 



Probable errors of clock corrections : 



March 29th = 0-018 sec. 



30th = 0-011 



31st = 0-024 



April 3rd = 0-009 



4th == 0-019 



From the observations in Table LIV, the rates of 991 have been formed for the 



* Astronomical date, Melbourne mean time. 

 74 



